Future Transformer for Long-term Action Anticipation
This addresses the problem of predicting distant future actions for real-world agents, offering a novel approach to improve accuracy and speed in video analysis.
The paper tackles long-term action anticipation by proposing Future Transformer (FUTR), an end-to-end attention model that predicts minutes-long future action sequences in parallel decoding, achieving state-of-the-art results on Breakfast and 50 Salads benchmarks.
The task of predicting future actions from a video is crucial for a real-world agent interacting with others. When anticipating actions in the distant future, we humans typically consider long-term relations over the whole sequence of actions, i.e., not only observed actions in the past but also potential actions in the future. In a similar spirit, we propose an end-to-end attention model for action anticipation, dubbed Future Transformer (FUTR), that leverages global attention over all input frames and output tokens to predict a minutes-long sequence of future actions. Unlike the previous autoregressive models, the proposed method learns to predict the whole sequence of future actions in parallel decoding, enabling more accurate and fast inference for long-term anticipation. We evaluate our method on two standard benchmarks for long-term action anticipation, Breakfast and 50 Salads, achieving state-of-the-art results.